Hearing Research Hearing Research 221 (2006) 44–58 www.elsevier.com/locate/heares Research paper Cochlear developmental defect and background-dependent hearing thresholds in the Jackson circler (jc) mutant mouse
Alfredo Calderon a, Adam Derr a, Barden B. Stagner b, Kenneth R. Johnson c, Glen Martin b, Konrad Noben-Trauth a,*
a Section on Neurogenetics, Laboratory of Molecular Biology, National Institute on Deafness and Other Communication Disorders, National Institutes of Health, 5 Research Court, Rockville, MD 20850, USA b Jerry Pettis Memorial Veterans Medical Center, Loma Linda, CA 92357, USA c The Jackson Laboratory, Bar Harbor, ME 04609, USA
Received 22 May 2006; received in revised form 19 July 2006; accepted 20 July 2006 Available online 7 September 2006
Abstract
Jackson circler (jc) is a spontaneous, recessive mouse mutation that results in circling behavior and an impaired acoustic startle response. In this study, we refined the phenotypic and genetic parameters of the original jc mutation and characterized a new mutant allele, jc2J. In open-field behavior tests, homozygous jc mutants exhibited abnormal circling and ambulatory behavior that was indis- tinguishable from that of phenotypically similar mutants with defects in the vestibule of the inner ear. The jc/jc and jc2J/jc2J mice had stable elevated auditory-evoked brainstem response (ABR) thresholds at the 16 kHz stimulus of 88 ± 9 dB sound pressure levels (SPL) and 43 ± 11 dB SPL, respectively. Peak latencies and peak time intervals were normal in jc mutants. The jc mice showed no measurable distortion-product otoacoustic emissions (DPOAEs) above the system noise floor. In the mutant cochlea, the apical turn failed to form due to the developmental growth arrest of the cochlear duct at the level of the first turn at gestational day 13.5. In a large intrasubspecific intercross, jc localized to a 0.2cM interval at position 25cM on chromosome 10, which is homologous to the human 6q21 region. On CZECHII/Ei and CAST/Ei backgrounds jc/jc mutant hearing thresholds at the 16 kHz stimulus were sig- nificantly lower than those observed on the C57BL/6J background, with means of 62 ± 22 dB SPL and 55 ± 18 dB SPL, respectively. Genome-wide linkage scans of backcross, intercross, and congenic progeny revealed a complex pattern of genetic and stochastic effects. 2006 Elsevier B.V. All rights reserved.
Keywords: Jackson circler; Hearing loss; Vestibular deficits; Cochlea malformation; Genetic background
1. Introduction The mutation was initially linked to the Steel locus on link- age group IV (Southard, 1970) and was placed at position The recessive Jackson circler (jc) mutation arose sponta- 32cM on chromosome 10 near marker D10Mit30 of the neously in the C57BL/6 mouse strain at The Jackson Lab- Mouse Genome Informatics consensus map (MGD, oratory in 1963 (Southard, 1970). Homozygous mutants 2005). Allelism tests with the closely linked deafness muta- were recognized by their erratic circling behavior and the tions waltzer (Cdh23v) and Ames waltzer (Pcdh25av) proved absence of a startle reflex at three to five weeks of age. negative (MGD, 2005). Recent studies showed that muta- tions underlying a deafness/waltzing phenotype occur pref- erentially in genes encoding proteins that are targeted to Abbreviations: ABR, auditory brainstem response analysis; DPOAE, stereocilia where they regulate organization, structure, distortion-product otocacoustic emission; QTL, quantitative trait locus * Corresponding author. Tel.: +1 301 402 4223; fax: +1 301 435 4040. growth and function of the hair bundle (Hasson et al., E-mail address: [email protected] (K. Noben-Trauth). 1995; Mburu et al., 2003; Rzadzinska et al., 2004; Siemens
0378-5955/$ - see front matter 2006 Elsevier B.V. All rights reserved. doi:10.1016/j.heares.2006.07.008 A. Calderon et al. / Hearing Research 221 (2006) 44–58 45 et al., 2002; Steel and Kros, 2001; Zheng et al., 2000). In The wild-derived CAST/Ei and CZECHII/Ei strains these mutants the stereocilia defects become apparent in were used in the mapping crosses to obtain a high degree the postnatal period coinciding with the functional matura- of heterogeneity at marker loci and to introduce pheno- tion of the hair cells. Mutations affecting the formation and typic diversity in the cross. To map the jc mutation an patterning of the cochlea however manifest during late intercross scheme was chosen to obtain the optimal num- embryogenesis, and also affect other developmental path- ber of meioses per offspring. To map genetic modifiers ways most notably the neural tube (Cordes and Barsh, alternative crosses and schemes were used to test for exper- 1994; Curtin et al., 2003; Montcouquiol et al., 2003; Phipp- imental reproducibility and robustness of statistical signif- ard et al., 1999; Quint and Steel, 2003). The isolated and icance levels. C57BL/6J-jc/jc males and females were pronounced circling phenotype of Jackson circler mice sug- crossed with CZECHII/Ei and offspring were intercrossed gested that the jc gene is critical for hair cells during post- to produce the B6/CZECH-F2 population. To generate natal growth. B6/CAST-F2 progeny, C57BL/6J-jc/jc males were crossed Phenotypic characteristics of a trait are often influ- to CAST/Ei females and the resulting F1 hybrids were enced by allelic variants present in the genetic background intercrossed. To obtain B6/CZECH-N2 offspring, (Nadeau, 2001). Hearing screens in mouse strains, con- C57BL/6J-jc/jc males were crossed with CZECHII/Ei ducted by measuring auditory-brainstem response thresh- females, and female F1 hybrid females were backcrossed olds and distortion-product otoacoustic emissions, have to C57BL/6J-jc/jc males. For serial backcrossing, B6/ revealed substantial variation within and among strains CZECH-jc/jc males were crossed to C57BL/6J+/jc females. with respect to onset and progression of hearing loss All animal studies followed the guidelines of the National (Jimenez et al., 1999; Zheng et al., 1999). Quantitative Institutes of Health and were approved by institutional trait loci underlying this phenotypic variance could be review boards. localized in some inbred strains (for review see (Johnson et al., 2006)). Intra-strain variation, such as that seen in 2.2. Open-field behavior C57BL/6J more than one year of age and in CBA/CaJ mice around two years of age is thought to result from Movement and activity behavior were measured using the breakdown of homeostatic mechanisms and environ- an Opto-Varimex 3 Activity Monitor from Columbus mental effects (Hequembourg and Liberman, 2001; Keith- Instruments (Columbus, OH, USA). Briefly, each mouse ley et al., 2004). Epistatic interactions and genetic was placed in a 7.5 in. · 8 in. · 12 in. (W · H · L) cage, sit- modifiers altering the disease outcome of pathological uated in a 17 in. · 17 in. · 8 in. open-field apparatus hearing loss alleles introduce additional variation (Ikeda equipped with one row each of horizontal and vertical sen- et al., 1999; Nadeau, 2003; Noben-Trauth et al., 1997). sors, with each row consisting of 15 infrared beams, such Methodical analyses of the background variation identi- that the horizontal and vertical beams were perpendicular fied new alleles linked with hearing loss (Johnson et al., to one another. The beams were one inch apart and 2006) and it is thought that such hypomorphic alleles 3 mm in diameter. Movements were tracked for 60 min and a combination thereof contribute to non-Mendelian and each test was repeated at least three times. Data were forms of hearing impairment. collected and analyzed using the Auto-Track v3.4 from In this work, we characterize different aspects of the Columbus Instruments. Ambulatory time (AT) was the inner ear phenotype of the Jackson circler mutant, describe time (measured in seconds, s) the animal exceeded the min- a new allele, present a high-resolution genetic map, and imal field. The x- and y-beam defined minimal field was report on the effect of the genetic background on the hear- 5.8 cm2. Clockwise (CR) and counter-clockwise (CCR) ing threshold distribution. rotations were defined as four consecutive right-handed or left-handed changes in direction, respectively. Two com- 2. Materials and methods mon laboratory strains (C57BL/6J and C3HeB/FeJ) and three wild-derived strains (CAST/Ei, CZECHII/Ei and 2.1. Mice MOLF/Ei) were included in the study to serve as genetic background controls for the various mutant alleles and All strains used in this study were obtained from The to ascertain the variability among inbred strains. Jackson Laboratory (Bar Harbor, ME, USA). The strain carrying the original jc mutation is designated C57BL/6J- 2.3. Auditory-evoked brain stem response (ABR) analyses jc/J (STOCK #563). The jc2J mutation arose spontane- ously in the B6.129S6-Il6tm1Kopf colony (STOCK #2650) ABR testing was performed as previously described and was given the official strain name B6(129S6)-jc2J/J (Zheng et al., 1999). All mice were presented with click, (STOCK #5292). Selective breeding against segregation 8-, 16-, and 32-kHz stimuli at intensities varying between of the Il6tm1Kopf allele (located on chromosome 11) elimi- 10 and 100 dB SPL. Each stimulus was presented 350 times nated the targeted mutation from the jc2J stock. Homozy- at a rate of 19 sÀ1 using a Blackman filter. Filter settings gous jc2J mutants show the typical circling behavior and used 100 Hz for high-pass and 3 kHz for low-pass. Animals are able to swim. were anesthetized using tribromethanol. ABR measurements 46 A. Calderon et al. / Hearing Research 221 (2006) 44–58 were taken for offspring of C57BL/6 J-+/jc · C57BL/6J-jc/ mary-tone levels were applied consisting of either jc matings. Unless otherwise indicated, ABR thresholds L1 = L2 = 55, 65 or 75 dB SPL. These levels were selected were given as the mean ± one standard deviation. The because they have been shown to be effective in screening mean ages of mice at the time of weaning were 87 ± 5 days for cochlear defects in mice (Jimenez et al., 1999; Jimenez for the (CAST/Ei · C57BL/6J-jc/jc)F1· F1 intercross et al., 2001). All mice were tested while anesthetized with progeny, and 47 ± 7 days for the (CZECHII/Ei · C57BL/ 1.5% concentration of isoflurane administered via a stan- 6J-jc/jc)F1· F1 intercross progeny, and 49 ± 8 days for dard veterinarian inhalation anesthesia machine. the (CZECHII/Ei · C57BL/6J-jc/jc) · C57BL/6J-jc/jc backcross progeny. 2.5. Paint fillings
2.4. Distortion-product otoacoustic emission measurements The bony labyrinth of the inner ear at E13.5 and E15.5 (day of gestation, E) was paint-injected following the pro- Distortion-product otoacoustic emissions (DPOAEs) tocol described by Martin and Swanson (1993). Briefly, the were measured in five C57BL/6J-jc/jc and five C57BL/6J- heads were fixed in Bodian fixative (5% glacial acetic acid; +/jc mice at 3 months of age. The 2f1 À f2 DPOAE mea- 1.85% formaldehyde; 75% ethanol) for 12 h, dehydrated in surements were performed as previously described for mice a graded series of ethanol (75%; 95%; 100%), and cleared in (Jimenez et al., 1999; Vazquez et al., 2001). Briefly, DPO- methyl-salicylate (EM Science). The heads were bi-dis- AEs were produced by the simultaneous presentation of sected, temporal brain tissue was removed and white latex two pure tones at the f1 and f2 primary frequencies (f2/f1 paint (0.025% in methyl-salicylate) was injected into one of ratio = 1.25) and levels (L1, L2) specified below. The pri- the ampulla. Pictures were taken on a Zeiss dissecting mary tones were generated by a dual-channel synthesizer microscope using a DXM1200 digital camera (Nikon). (model 3326A, Hewlett-Packard, Palo Alto CA) and atten- Embryos were obtained from C57BL/6J-+jc · C57BL/6J- uated under the control of a personal computer system jc/jc and C57BL/6J-jc/jc · C57BL/6J-jc/jc matings. The using customized software. The f1 and f2 primaries were time at which a maternal plug was observed was set as presented over two separate ear speakers (Realistic Dual E0.5. Radial Horn Tweeters, Tandy Corp, Dallas TX) and deliv- ered through a commercially available acoustic probe 2.6. Crosses and genetic mapping incorporating a calibrated microphone assembly (ER- 10B+, Etymotic Research Elk Village, IL) that was Reciprocal matings between C57BL/6J-jc/jc and wild- inserted snugly into the outer ear canal. The two primary derived inbred CZECHII/Ei mice (Mus musculus musculus) tones were allowed to mix acoustically in the ear canal to produced the B6/CZECH-+/jc F1 generation, which were avoid artifactual distortion. intercrossed to generate 990 F2 offspring. F2 mice were Ear-canal sound pressure was sampled and averaged phenotypically classified based upon their vestibular phe- using two methods. For f2 frequencies from 6.3 to notype and their hearing thresholds. Non-informative 22.5 kHz, sampling and averaging were performed by a recombinants were crossed to C57BL/6J-+/jc heterozygotes digital signal processor (DSP) in the microcomputer. and offspring with informative genotypes were tested for DPOAE levels were measured automatically over the 92- hearing function by ABR analysis. D10Mit markers were ms duration of the primary tones from the amplitude spec- obtained from Research Genetics (Invitrogen, Carlsbad, trum resulting from a 4096-point (bandwidth = 10.8 Hz) CA). D10Ntra (simple-sequence-length and single-nucleo- fast Fourier transform (FFT) of four synchronously aver- tide) polymorphic markers were developed in our labora- aged samples. To determine the corresponding noise floors tory. Primer sequences are available upon request. (NFs), the levels for the ear-canal sound pressure for five Genetic distances were computed using MapManager FFT frequency bins above and below the DPOAE-fre- QTX software (Manly et al., 2001). The jc physical interval quency bin (i.e., ±54 Hz) were averaged. was analyzed using the Human Genome Browser at For f2 frequencies from 21.5 to 54.2 kHz, a computer- www.genome.cse.ucsc.edu (assembly May 2005). Chromo- controlled dynamic-signal analyzer (3561A, Hewlett-Pack- somal locations were obtained from the Mouse Genome ard, Palo Alto CA) was used. Within this higher-frequency Database at www.informatics.jax.org (Eppig et al., 2005). test range the DPOAE levels were based upon the results of four spectral averages of the ear-canal signal. For this 2.7. Genotyping method NFs were estimated by averaging the levels of the ear-canal sound pressure for the two FFT frequency Genomic DNA was extracted from tail clips using bins below the DPOAE frequency (i.e., for 3.75 Hz below DNeasy (Qiagen, Valencia, CA). For amplification 20– the DPOAE). Using either the DSP- or signal-analyzer sys- 50 ng of extracted DNA served as the template in a 10 ll tems, no artifactual DPOAEs were ever measured in a reaction containing 0.2 lM each primer, 200 lM dNTP, hard-walled cavity that approximated the size of the mouse 10 mM Tris–HCl, pH 8.3, 50 mM KCl, 1.5 mM MgCl2, outer ear canal. DPOAEs were measured as DP-grams, i.e., and 0.25 U AmpliTaq DNA polymerase (Applied Biosys- DPOAE levels as a function of f2 frequency. Three pri- tems). The reactions were incubated at 95 C for 1 min, A. Calderon et al. / Hearing Research 221 (2006) 44–58 47 then cycled 50 times through 95 C for 45 s, 55 C for mice (n = 22), six weeks of age and older, spent an average 1 min, 72 C for 1 min, followed by a final extension at (± one standard deviation) of 1480 ± 540 seconds per hour 72 C for 10 min and cooled to 4 C for 2 min. For capil- (s/h) in non-stereotypic behavior compared to 576 ± 238 s/ lary electrophoresis, 1 ll of amplified product was diluted h of littermate controls (n =7)(Fig. 1A; Table 1). Mice of 1:10 in formamide containing 0.1 ll GenScan 500 the C57BL/6J and C3HeB/FeJ strains spent an average of ROX size standard, separated on a 3730 DNA Analyzer 676 ± 309 s/h and 490 ± 173 s/h in ambulation, which was (Applied Biosystems), and analyzed using GenMapper not different from the behavior of +/jc heterozygotes software v3.1 (Applied Biosystems). For agarose gel elec- (p > 0.05, Dunnet’s multiple comparisons test). Hyperac- trophoresis, PCR product was loaded on 4% Metaphor tive behavior of jc/jc mutants was also significantly differ- 1% NuSieve 3:1, separated and visualized with ethidium ent from activities of the wild-derived CZECHII/Ei and bromide staining. CAST/Ei strains, which spent an average of 325 ± 119 s/ h and 826 ± 315 s/h in ambulation, respectively (p < 0.01, 2.8. Genome-wide linkage scan Dunnet’s multiple comparisons test). As a general observation, the behavioral phenotype Microsatellite markers that are polymorphic between became apparent around weaning age, but full expression C57BL/6J, and CAST/Ei mice were selected from the Cen- varied among individual mutants. The mean ambulation ter for Inherited Disease Research website (www.cidr.jh- time in three-week old mutants was 684 s/h and had mi.edu/mouse/mouse_strp.html)(Witmer et al., 2003) and increased to 1321 s/h at six weeks of age (Fig. 1B, Table purchased from Applied Biosystems, Invitrogen or IDT. 1). Due to the wide scatter of data points there was no sig- For the genome-wide scan on the B6/CZECH-jc/jc N2 nificant difference between the means of different age population, 42 jc/jc progeny were genotyped at 137 marker groups (p > 0.05, ANOVA). Ambulation behavior of jc/jc loci. For the analyses of the B6/CAST-jc/jc F2 cross, a mice was compared with that of the phenotypically similar selective genotyping protocol was used. Twenty progeny deaf waltzing mutants Cdh23v6J/v6J, Cdh23v2J/v2J, from each tail of the hearing threshold distribution (40 Pcdh15nmf19/nmf19, Espnje/je, Myo6sv/sv, Myo15Ash2/sh2,and F2 jc/jc homozygotes total), containing most of the geno- Mcoln3+/Va. While there were clear differences between typic information, were typed at 127 markers. For the scan the mutants and the respective heterozygous or wildtype on the B6/CZECH-jc/jc F2 cross, 45 jc/jc progeny were control group, the means of the mutant strains, including typed at 143 markers. The linear regression method of jc, were not significantly different from one another MapManager QTXb20 was used to analyze association (p > 0.05; ANOVA) (Table 1). of genotype data with trait values (Manly et al., 2001). A significant portion of the ambulatory behavior seemed Empirical thresholds to estimate genome-wide significance to be correlated with extreme circling movements. Homo- of likelihood-ratio-statistic (LRS) scores were computed by zygous jc/jc mice (n = 22) performed 397 ± 464 clockwise performing 1000 permutations (Churchill and Doerge, (CRs) and 478 ± 675 counter-clockwise (CCRs) rotations 1994). Genome-wide p values of 0.63 and 0.05 were taken per hour compared to 50 ± 48 CR and 45 ± 41 CRR in as the suggestive and significant linkage levels, respectively heterozygotes (n = 7). There was no significant difference (Lander and Kruglyak, 1995). Pair-wise marker regression in the direction of the circling motion among individual analyses were performed at p =10À5 and the significance mutant mice (p > 0.05) (Fig. 1C). The combined CRs and level for the interaction was computed by permutation test- CCRs ranged from 0 to 6255 rotations per hour. Ambula- ing. LRS scores were converted to LOD scores by dividing tion and circling behavior showed a significant correlation the LRS value by the factor 2ln(10). (p < 0.0001) with a mean coefficient of determination (r2, Pearson) of 0.45 with upper and lower 95% confidence 2.9. Statistical analyses intervals of 0.17 and 0.73, respectively (n =5)(Fig. 1D). Similar to the ambulation phenotype, there was no signifi- Descriptive statistics, analyses of variance (ANOVAs), cant difference in circling behavior between jc/jc and other Dunn’s correction for multiple testing, and normality tests deaf waltzing mutants (Table 1). Finally, jc mutants cov- were performed using GraphPad Prism 4.0b software (San ered a distance of 409 ± 125 m (maximum = 806 m) during Diego, CA). a one-hour test, which deviated significantly from heterozy- gous littermates and normal inbred strains (Fig. 1E). 3. Results 3.2. Auditory brain stem responses in jc and jc2J mice 3.1. Hyperactivity and circling behavior ABR measurements revealed that homozygous jc mice To assess the vestibular deficits in Jackson circler had measurable responses to click and pure tone stimuli, mutants, we measured movement behavior in an open-field although at high sound pressure levels only. Two-week old test system. During a 60-min test period, homozygotes jc/jc mutants (n = 12) had a mean hearing threshold of spent significantly more time in ambulation than their het- 83 ± 11 dB SPL to the click stimulus; individual thresholds erozygous littermates (p < 0.0001; t-test). Specifically, jc/jc ranged from 65 dB SPL to 100 dB SPL (Table 2). The 48 A. Calderon et al. / Hearing Research 221 (2006) 44–58
Fig. 1. Open field motor behavior of the jc mouse. (A) A histogram of ambulation time of C57BL/6J-+/jc (grey bars) and C57BL/6J-jc/jc (black bars) mice is shown. From the heterozygous group, 7 animals six weeks of age or older were tested in a total of 29 tests. From the homozygous group, 22 animals 6 weeks of age or older were tested in a total of 114 tests. The black curves represent the best-fit of a normal distribution and the R2 values, which indicate the percentage of variation accounted for by the model, are given. (B) The distribution of ambulation time of C57BL/6J-jc/jc mice (y-axis) as function of age (x-axis) is summarized. Each circle represents the value obtained for an individual animal and the horizontal line represents the median of the distribution. (C) The distributions of the numbers of clock-wise (CR) and counter-clock wise (CCR) rotations of C57BL/6J-jc/jc and C57BL/6J-+/jc mice are shown as box-and-whisker blots, indicating the minimal and maximal value, the 25th and 75th percentile values and the median. Letters annotate significant differences (p < 0.05). Test groups were as in A. (D) The correlation between ambulation time (x-axis) and number of rotations (y-axis) is shown for one representative C57BL/6J-jc/jc homozygote. The straight line shows the regression curve with the R2 value as the best-fit parameter. (E) The mean distances traveled by the C57BL/6J-jc/jc (test group as in (A)), C57BL/6J-+/jc (test group as in (A)), and C57BL/6J (B6, 45 tests) groups of mice are presented. Distribution of values is given in box-and-whisker plot format. mutants also exhibited elevated thresholds to 8 kHz, 16 kHz tests). Hearing impairment did not proceed to a complete and 32 kHz pure-tone stimuli, and the thresholds increased hearing loss; instead mutants at 22 weeks of age and older with increasing frequency. The thresholds varied consider- had a mean hearing threshold of 93 ± 6 dB SPL (n = 18). ably among the jc/jc mutant mice with standard deviations The presence of measurable thresholds in the mutants ranging from 7 to 12 dB SPL. Comparison of the threshold allowed us to determine the latency peak response and means from mutants at 3- to 12-weeks of age did not reveal inter-peak intervals for peaks I–IV at 90, 95 and 100 dB any age-dependent changes in ABR thresholds over the 10- SPL. Following a 16 kHz tone burst at 100 dB SPL, the week test period (p > 0.05; Dunnet’s multiple comparison first peak (I) appeared with a delay of 1.6 ± 0.13 ms in A. Calderon et al. / Hearing Research 221 (2006) 44–58 49
Table 1 Movement behavior in jc, common inbred strains, and circling mutants Strains Age Ambulation (s) Distance (m) Rotations N/T Mean SD Mean SD Mean SD C57BL/6J 12–20 676.2 309.0 136.2 52.7 87.5 63.3 4/45 C3HeB/FeJ 8–16 490.2 172.6 105.1 39.3 129.1 240.9 2/22 MOLF/Ei 12 481.6 143.5 111.6 41.0 64.1 54.1 2/10 CZECHII/Ei 12 325.5 119.1 77.6 31.0 47.8 33.5 2/12 CAST/Ei 4–8 826.3 314.5 215.5 113.6 152.6 94.2 4/10 C57BL/6J-jc/jc 3 684.8 596.7 167.3 165.4 143.3 189.7 12/12 4 940.3 428.9 224.8 127.4 427.4 580.7 12/12 5 1073.0 614.2 270.1 190.8 409.5 413.0 12/12 6 1321.0 704.3 323.9 195.8 515.3 524.5 12/12 7 1312.0 558.9 311.3 154.7 623.8 563.8 12/12 8 1471.0 643.2 358.8 186.6 672.8 757.5 12/12 9 1369.0 549.5 335.9 161.5 987.0 1420.0 11/11 10 1400.0 665.1 345.9 203.2 1032.0 1901.0 11/11 11 1350.0 623.3 305.9 152.0 1858.0 1976.0 8/8 12 1392.0 585.3 323.9 169.1 621.3 587.9 7/7 C57BL/6J-+/jc 3 344.9 175.9 95.7 65.8 42.7 20.5 4/4 4 460.9 117.3 94.8 20.0 89.7 90.2 4/4 5 504.5 76.1 120.3 7.0 75.2 15.2 4/4 6 690.2 260.7 161.4 53.4 160.5 86.4 4/4 7 546.6 210.3 117.2 48.6 54.0 34.4 4/4 8 550.7 92.7 120.1 33.7 70.7 29.6 4/4 9 506.6 69.6 106.5 17.1 62.3 26.9 3/3 10 513.2 202.5 109.8 43.5 74.0 34.1 3/3 11 710.0 318.7 143.6 81.7 37.7 31.1 3/3 12 834.5 192.8 200.7 64.6 156.0 101.8 2/2 Cdh23v6J/v6J 4–16 1811.0 405.8 451.0 124.4 1164.0 960.9 23/59 Cdh23+/v6J 4–16 580.5 167.5 129.4 44.2 102.2 58.9 23/46 Cdh23v2J/v2J 4–16 1675.0 468.6 405.1 14.0 838.5 643.2 18/53 Cdh23+/v2J 4–16 730.5 277.6 145.7 53.1 145.0 136.3 13/38 Myo6sv/sv 36 1343.0 487.3 343.1 147.1 675.1 513.3 1/7 Myo6+/sv 36 533.3 168.8 114.6 37.6 83.4 40.6 1/7 Pcdh15nmf19/nmf19 16 1887.0 291.9 466.3 99.6 1107.0 599.4 3/17 Pcdh15+/nmf19 16 717.7 255.8 152.0 55.6 191.7 239.1 5/31 Espnje/je 16–24 1548.0 870.6 344.7 68.6 669.6 596.5 4/36 Espn+/je 16–24 579.5 277.5 107.6 54.3 108.5 117.7 6/35 Mcoln3+/VaJ 8–12 535.1 169.3 115.5 36.7 63.2 40.7 4/10 Mcoln3+/Va 48 1812.0 374.6 450.1 127.7 1147.0 773.4 14/49 Mcoln3+/+ 48 499.0 137.0 102.9 31.9 99.4 170.8 8/25 C57BL/6J-jc/jc 3–12 1575.0 486.5 409.0 125.5 985.6 1051.0 20/112 C57BL/6J-+/jc 3–12 602.9 239.6 131.6 54.3 116.8 126.6 12/83 Myo15a?/sh2 4–12 741.6 291.2 167.4 77.2 244.6 411.7 5/25 Myo15ash2/sh2 4–12 1761.0 317.6 474.7 138.1 941.9 384.5 5/38 Age is given in weeks; s, seconds; SD, standard deviation; m, meter; N/T, number of mice used for number of tests.
mutants (n = 9) and 1.6 ± 0.03 ms in heterozygotes (n = 6). ing loss in jc2J was not progressive, but rather their hearing With decreasing sound pressure levels, the latencies thresholds remained stable at 40 dB SPL (16 kHz stimulus) increased while the inter-peak intervals remained constant. over the 19-week test period (Table 2). There was no statistically significant difference in peak latencies between the mutants and age-matched heterozy- 3.3. Distortion-product otoacoustic emissions (DPOAEs) in gous controls (Fig. 2, Table 3). jc mutants The jc2J homozygotes (n = 4) also exhibited elevated hearing thresholds, with the mean threshold for the To differentiate the cause of hearing loss, DPOAE func- 16 kHz stimulus being 43 ± 11 dB SPL (Table 2). At the tion was determined in a cohort of five +/jc and five jc/jc 8 kHz and 16 kHz frequencies, jc2J mutants had signifi- mice at 12-weeks of age. The DPOAEs of heterozygous cantly better hearing than jc/jc mice (p < 0.05 and mice fell within the normal range for all three primary tone p < 0.001, respectively, Bonferroni multiple comparison levels (55 dB SPL, 65 db SPL, and 75 dB SPL). In contrast, test), but at the click and 32 kHz stimulus hearing was jc homozygotes had no measurable DPOAEs above the equally poor (p > 0.05). Similar to that in the jc mice, hear- noise floor at these primary-tone levels (Fig. 3). 50 A. Calderon et al. / Hearing Research 221 (2006) 44–58
Table 2 ABR thresholds in jc and jc2J mutants Cross Age Click 8 kHz 16 kHz 32 kHz N jc/jc 2 83 ± 11 76 ± 12 90 ± 7 97 ± 5 12 +/jc 236±220±517±335±310 jc/jc 3 90 ± 12 83 ± 14 91 ± 5 96 ± 6 10 +/jc 3 35 ± 5 24 ± 10 17 ± 7 42 ± 12 12 jc/jc 4 86 ± 10 73 ± 11 83 ± 10 90 ± 7 25 +/jc 432±325±617±64±610 jc/jc 8 96 ± 5 81 ± 13 86 ± 8 95 ± 6 15 +/jc 832±323±212±338±87 jc/jc 12 95 ± 6 96 ± 12 86 ± 11 86 ± 8 15 +/jc 12 33 ± 6 25 ± 4 12 ± 5 34 ± 2 6 jc/jc >22 94 ± 6 87 ± 14 93 ± 6 97 ± 4 18 +/jc >22 39 ± 2 39 ± 2 26 ± 10 52 ± 4 4 jc2J/jc2J 6 76 ± 7 67 ± 15 43 ± 11 85 ± 15 4 +/jc2J 642±437±415±057±42 jc2J/jc2J 19 68 ± 9 65 ± 6 41 ± 6 70 ± 0 4 Age is given in weeks; data represent means ± SD; kHz, kilohertz; N, number of animals tested.
heterozygotes the cochlea outgrowth developed normally completing the first turn by E15.5 (Fig. 4A and D). In jc homozygotes the cochlea duct initially seemed to form nor- mally but at around E13.5 signs of a growth arrest were noticeable (Fig. 4B and C). The truncation became more apparent by E15.5 and an aberrant structure had formed at the apex (Fig. 4E and F). Twenty-six cochleae from 15 jc/jc embryos were analyzed and all showed the develop- mental arrest. Only negligible inter-individual variation in expressivity was observed among the mutants and no later- ality effect was noted. In cochleae of 14-day old mutant neonates the apical turn was absent (Fig. 4H and I). No discernable gross abnormalities in the anatomy of the ves- tibule were seen.
3.5. High-resolution genetic map of jc
With the goal of eventually identifying the jc gene, the Fig. 2. Auditory-evoked brain stem recording (ABR) waveforms in jc jc locus was mapped with high resolution. A total of 1980 mutants. Representative ABRs following a 16 kHz tone-burst at various chromosomes from an intrasubspecific intercross between sound pressure levels (y-axis) in jc/jc and +/jc mice are shown. The C57BL/6J-jc/jc and CZECHII/Ei were analyzed with a series locations of peaks I–V are indicated. Scale bars of the ABRs are given. dB SPL, decibel sound pressure level; ms, millisecond; lV, microvolt. of polymorphic markers located within the centromeric half of chromosome 10. Linkage was found between D10Mit108 3.4. Developmental arrest of cochlear growth and D10Mit55 (v2 test; p < 0.001; df = 1). Twenty-one recombinant chromosomes were further characterized with The hearing impairment in the jc/jc mutants was accom- newly developed D10Ntra markers. The following marker panied by a structural aberration of the cochlear duct. In jc order and distances (centiMorgan ± standard error) were
Table 3 Peak latencies in +/jc and jc/jc mutants Cross dB SPL I II III IV V N +/jc 100 1.6 ± 0.03 2.4 ± 0.09 3.3 ± 0.10 4.3 ± 0.10 5.2 ± 0.08 6 95 1.7 ± 0.03 2.5 ± 0.09 3.3 ± 0.07 4.3 ± 0.12 5.2 ± 0.06 6 90 1.7 ± 0.03 2.5 ± 0.06 3.4 ± 0.08 4.4 ± 0.15 5.3 ± 0.10 6 jc/jc 100 1.6 ± 0.13 2.6 ± 0.42 3.6 ± 0.45 4.5 ± 0.48 5.4 ± 0.38 9 95 1.7 ± 0.14 2.7 ± 0.43 3.8 ± 0.44 4.6 ± 0.46 5.5 ± 0.34 9 90 2.1 ± 0.45 3.0 ± 0.68 4.0 ± 0.67 5.0 ± 0.72 5.5 ± 0.32 8 dB SPL, decibel sound pressure level; N, number of animals tested; data represent mean ± SD. A. Calderon et al. / Hearing Research 221 (2006) 44–58 51
Fig. 3. DPOAEs in jc mutants. Average DP-grams (n = 5 ears) collected at three primary-tone levels for +/jc (solid circles) and jc/jc mice (solid triangles) at 3 mo of age. DPOAEs from +/jc mice fell in the normal range (± 1SD) of DPOAEs (solid gray lines) obtained from CBA/CaJ mice known to have normal hearing. DPOAEs from jc/jc mice were absent and were at the noise floor of the recording system (dotted gray line). Error bars indicate ± 1SD. obtained: D10Mit108 – 0.05 ± 0.05 – (D10Ntra116, average age of eleven weeks a cohort of 67 C57BL/6J-jc/ D10Ntra117) – 0.05 ± 0.05 – D10Ntra12 – 0.1 ± 0.07 – jc mutants had a mean ABR threshold of 90 ± 9 dB SPL. (D10Ntra14, D10Ntra15, jc) – 0.1 ± 0.07 – (D10Ntra118, The lowest threshold measured was 60 dB SPL, but most D10Ntra119) – 0.51 ± 0.16 – D10Ntra111 – 0.05 ± 0.05 – of the thresholds (P25th percentile) were equal to or (D10Ntra114, D10Mit55)(Fig. 5). The jc minimal genetic higher than 85 dB SPL (Fig. 6A, Table 4). On both interval was defined by two proximal and two distal recom- CAST/Ei and CZECHII/Ei genetic backgrounds hearing binations and comprised an interval length of 0.2 ± 0.1cM. thresholds of jc/jc mutants were significantly lower than On the mouse physical map (NCBI m34, assembly May that on the C57BL/6J background (p < 0.01; Dunnet’s 2005), the closest recombinant markers D10Ntra2 and multiple comparison test). In particular, F2 homozygotes D10Ntra16 were separated by a genomic sequence approxi- (n = 80) from a (C57BL/6J-jc/jc · CAST/Ei) F1 intercross mately 170 kb in length. This region contained several and F2 homozygotes (n = 48) from a (C57BL/6J-jc/ expressed sequence tags, numerous exons predicted by vari- jc · CZECHII/Ei) F1 intercross had thresholds of ous gene-prediction algorithms, and multiple cross-species 55 ± 13 dB SPL and 62 ± 22 dB SPL, respectively. Hearing conserved sequences. The human chromosomal region that thresholds in both the B6/CAST-jcjc and B6/CZECH-jc/jc is homologous to the jc minimal genetic interval is present F2 populations followed a normal distribution (p = 0.8 and at 6q21. Craniometaphyseal dysplasia (OMIM 218400), p = 0.1, respectively; D’Agostino and Pearson test) which is a recessive syndrome characterized by cranial defor- (Fig. 6B and C). Hearing thresholds in jc homozygotes mities including hearing loss, is among several human condi- (n = 43) obtained from a [(C57BL/6J-jc/jc · CZECHII/ tions that have been linked to the 6q21 region (Iughetti et al., Ei) F1 · C57BL/6J-jc/jc] backcross were also significantly 2000). However marker D6S302, which defines the centro- lower (p < 0.01) than those of jc homozygotes on the iso- meric border of the linked interval, is located 4 Mb telomeric genic background. Thresholds, however, were not normally to the recombinant jc marker D10Ntra118 excluding this distributed (p = 0.0002), but instead showed a bi-modal syndrome as the human jc disease homolog. scatter, with the first tier comprising thresholds from 15 The jc2J mutation was mapped in a [B6(129S6)-jc2J/ to 50 dB SPL (n = 17) and the second tier containing jc2J · CAST/Ei]F1 · F1 intercross. Fifty-three homozy- thresholds from 60 to 95 dB SPL (n = 26). The ratio gous mutants were genotyped with microsatellite markers between these two groups was 1:1.5. (Fig. 6D). Interest- and the following map location was obtained: D10Mit80 ingly, the genetic background did not affect the degree of – 1.7cM – (D10Mit55, jc2J) – 3cM – D10Mit138 (Fig. 5). the cochlea dysplasia (data not shown). To test whether To formally test for allelism, heterozygous +/jc2J mice were thresholds would distribute into discrete groups over time, crossed with C57BL/6J-jc/jc mutants; three out of eight jc homozygotes from the normal hearing spectrum were animals produced by this cross showed the circling and tested at older ages. Consistent with the observation in hearing impairment phenotype that is characteristic of mice on the isogenic background, age had no effect on the jc mutant. the thresholds in these intercross progeny (data not shown). 3.6. ABR threshold distributions on different genetic backgrounds 3.7. Genome wide linkage scan
Hearing thresholds on the C57BL/6J coisogenic back- The lower thresholds observed in the progeny of the ground showed a characteristic variation, such that at an outcrosses and the differing distributions observed in the 52 A. Calderon et al. / Hearing Research 221 (2006) 44–58
Fig. 4. Cochlea malformation in jc mutants. Pictures (A)–(F) show paint-fills of the inner ear’s membranous labyrinth. Ears shown in (A)–(C) were obtained from E13.5 day old embryos and ears shown in (D)–(F) were from E15.5 day old embryos. Pictures (G)–(I) show inner ears from 14-day old neonates in which the temporal bone was removed to expose the cochlea duct. Two mutant ears are shown per time point. Ears shown in A, D, and G were obtained from +/jc mice and ears shown in B, C, E, F, H, and I were obtained from jc/jc mutants. To aid in phenotype classification ears from the progeny of jc/jc · jc/jc matings were included in the study. Arrowheads point to the upper cochlea turn. Asterisks indicate the missing apical turn. c, cochlea duct; s, saccule; u, utricle; es, endolymphatic sac, l, lateral; a, anterior, p, posterior semicircular canal; scale bar in G, H, and I equals 0.5 mm. progeny of the intercrosses and the backcross led to the tion, genome-wide linkage analyses were conducted. Forty- hypothesis that dominant alleles contributed by the CAST/ two B6/CZECH-jc/jc backcross progeny, representing 98% Ei and CZECHII/Ei strains partially restored hearing in of the jc/jc N2 population, were genotyped at 137 marker the outcrossed homozygous mutants. To test this proposi- loci. Marker regression analyses identified a significant A. Calderon et al. / Hearing Research 221 (2006) 44–58 53
Fig. 5. Genetic map and haplotype of the jc locus at chromosome 10. (A) The haplotype distribution in the F2 population is shown. Polymorphic markers are indicated on the left, numbers of recombinations per total meioses are given on the right and haplotype frequencies are shown at the bottom. Black boxes indicate alleles of the jc chromosome and the white boxes indicate the CZECHII/Ei-derived allele. (B) The marker locations on proximal chromosome 10 (MMU Chr10) are shown. Marker order is shown on the left and genetic distances are given on the right. The jc critical interval is highlighted. (C) The chromosomal location of the jc2J locus is shown. Markers are given on the left and marker distances (in cM) are given on the right.
QTL associated with marker D18Mit60 (position 16cM) and 16 kHz traits at D9Mit151 (position 72cM) with with LOD scores of 3.1 and 3.5 (significance threshold: LOD scores of 5.5 and 4.8, respectively (Table 5). The 3.1) for the 8 kHz and 16 kHz stimuli, respectively. Pair- B6/CAST-jc/jc F2 offspring were scanned using a selective wise marker regression analyses to test for interacting loci genotyping protocol. Twenty-one F2 progeny from the revealed no significant association (Table 5). Forty-five lower 625th percentile range (hearing threshold range: B6/CZECH-jc/jc F2 progeny (representing 94% of the jc/ 15 dB SPL – 45 dB SPL) and nineteen F2 progeny from jc F2 population) were genotyped at 143 marker loci. Sig- 68 to 97th percentile range (hearing threshold range: nificant genome-wide linkage was found with the 8 kHz 65 dB SPL – 95 dB SPL) were genotyped at 127 marker 54 A. Calderon et al. / Hearing Research 221 (2006) 44–58
analyses to test for interacting loci revealed no significant association.
3.8. Advanced backcross lines
To further test the hypothesis that dominant modifier alleles reduced hearing thresholds in the outcrossed homo- zygotes, two segregating lines were generated. Through repeated backcrossing of F2 and N2 B6/CZECH-jc/jc homozygotes with the lowest thresholds at the 16 kHz stimulus to C57BL/6J-+/jc heterozygotes we expected to isolate the CZECH-derived genomic segment(s) that are responsible for the lower thresholds. At generation N5, jc/jc homozygotes with the lowest (30 dB SPL at 16 kHz) and highest thresholds (95 dB SPL) from each of the two independent backcross lines were genotyped at 137 marker loci to identify common CZECH-derived chromosomal segments. As expected each of the two lines segregated a unique set of genomic segments (six and eight) from the donor strain; segments from chromosomes 2, 3, and 11 were segregating in both lines. Additional N5 progeny from both lines (n = 28, Fig. 7) were genotyped to correlate the congenic intervals with hearing status. None of the con- genic intervals from either line showed a clear correlation with hearing status (ANOVA) and appeared to be ran- domly assorted among the normal and hearing-impaired N5 jc/jc homozygotes. We next backcrossed a jc/jc homozygote of the N7 gen- eration with a 100 dB SPL threshold (not expected to carry modifying CZECH-alleles) with a C57BL/6J-+/jc mouse. Unexpectedly, we recovered jc/jc homozygotes with hear- ing thresholds in the 40–50 dB SPL range. The hearing thresholds for this N8 control group did not differ from that of the test groups of the N3–N7 generations (ANOVA; p > 0.05) (Fig. 7).
4. Discussion
4.1. Cochlea malformation Fig. 6. ABR threshold distributions of jc/jc homozygotes on different genetic backgrounds. Shown in (A) is the histogram of hearing thresholds In this study, we identified several features that are char- for the 16 kHz tone-burst stimulus in C57BL/6J-+/jc (grey bars) and acteristic of the Jackson circler phenotype. Perhaps the C57BL/6J-jc/jc (black bars) mice. Shown in B and C are the histograms of most interesting finding was the observation that jc hearing threshold distributions of jc/jc homozygotes from the (CAST/ mutants showed a response, albeit abnormal, to acoustic Ei · C57BL/6J-jc/jc)F1· F1 intercross and the (CZECHII/Ei · C57BL/ 6J-jc/jc)F1· F1 intercross, respectively. The best-fit curves for a Gaussian stimuli at a young age and preserved this residual hearing distribution are shown as solid black curves. The regression coefficient for over an extended period of time. This hearing retention is the curve is given as R2. Shown in D is the histogram of hearing thresholds in contrast to the phenotypes observed for many spontane- of jc/jc homozygotes from the (CZECHII/Ei · C57BL/6J-jc/jc) · C57BL/ ous deafness waltzing mutants, which are either congeni- 6J-jc/jc backcross. For (A)–(D) the horizontal axis denotes hearing tally deaf or show a rapid decline in hearing function thresholds in decibel sound pressure levels (dB SPL) and the vertical axis gives the number of animals. after birth. The hearing at high sound pressure levels in coisogenic jc mutants was characterized by a scattering of thresholds at all stimuli, which suggests that cochlea func- tion in jc is prone to stochastic effects. The absence of loci. One marker on chr 12 (D12Mit172) showed suggestive DPOAEs at the primary-tone levels of 55, 65, and 75 dB linkage with a LOD score of 5.0 (significance threshold: SPL was consistent with the absence of ABR thresholds 2.9) at the 16 kHz stimulus. The click, 8 kHz and 32 kHz at the corresponding stimuli and suggests that dysfunction stimuli showed no linkage. Pair-wise marker regression of outer hair cells may underlie the hearing deficit. The A. Calderon et al. / Hearing Research 221 (2006) 44–58 55
Table 4 ABR thresholds of jc/jc homozygotes Cross Click 8 kHz 16 kHz 32 kHz C57BL/6J-jc/jc N 69 69 69 68 Mean 93.5 85.1 89.4 97.0 SD 8.3 13.1 9.3 5.6 C57BL/6J-+/jc N 32 32 32 32 Mean 33.9 25.8 21.1 37.7 SD 7.3 8.7 8.3 7.8 B6/CZECH-jc/jc F2 N 49 49 48 44 Mean 72.4 58.8 61.8 79.9 SD 11.3 21.1 22.0 16.1 B6/CAST-jc/jc F2 N 82 81 80 74 Mean 70.4 54.2 55.1 75.6 Fig. 7. ABR threshold distribution as a function of backcross generation. SD 12.6 15.4 18.3 14.3 Hearing thresholds for the 16 kHz stimulus in jc/jc homozygous backcross B6/CZECH-jc/jc N2 N 44 43 43 43 progeny at the indicated backcross generation (N2–N7) are shown. Each Mean 75.3 55.8 61.3 78.1 circle represents the threshold in dB SPL of one animal. The horizontal SD 14.3 21.0 24.7 20.2 black line indicates the mean of the population. The threshold means in the advanced backcross lines (generation N3–N7) ranged from 71 dB SPL N, number of animals tested; SD, standard deviation; kHz, kilohertz. to 80 dB SPL and did not differ significantly (p > 0.05; Bonferroni’s multiple comparison test). N8 ctr, control group at generation N8. residual hearing may be mediated by still functioning inner hair cells. The comparison of ABR thresholds showed that res et al., 1996; Wang et al., 2005; Xu et al., 1999). jc2J mutants had significantly better hearing than jc homo- Although the deafness waltzing phenotype initially seemed zygotes of a similar age. Thus it may be that the jc2J allele is to be indicative of a type of degenerative process within the less severe than the jc allele. organ of Corti, the paint-filling data suggest the jc gene has ABR waves are thought to originate in the cochlea (peak a function in the growth and extension of the cochlear coil I) and from brainstem nuclei including the cochlea nucleus during development. (peaks II and III), medial superior olivary nuclei (peak IV), The cochlea malformation seen in the jc mutant mouse and nuclei in the inferior colliculus (peak V) (Melcher and is reminiscent of Mondini deformity in humans, which Kiang, 1996a; Melcher et al., 1996b; Melcher et al., 1996c). was first described by Carlo Mondini in 1791 who reported Myelin-deficiency of auditory nerve fibers in trembler mice missing apical turns in both cochleae of a deaf boy (Mon- (Pmp22TrÀJ) were recently shown to result in a prolonged dini, 1997; Mundinus, 1791). Nowadays, Mondoni dyspla- peak I latency (Zhou et al., 1995a; Zhou et al., 1995b). sia comprises a spectrum of morphological aberrations in The peak-latencies measured in the jc mutants were compa- the inner ear that include aplastic cochlea, enlarged vestib- rable to that seen in the heterozygotes, indicating that the ular aqueduct (EVA) and other deformations of the bony main generators of the ABR were functioning properly. and membranous labyrinth. Clinical and molecular studies The peak latency of 1.7 ± 0.03 ms seen with 90 dB SPL correlated EVA and Mondini dysplasia with Pendred syn- stimuli in +/jc heterozygotes also correlated well with the drome (MIM 274600) and mutations of the pendrin gene 1.6 ms latency recorded in one-month-old C57BL/6J mice SLC26A4 (Campbell et al., 2001; Pryor et al., 2005; Tsu- (Hunter and Willott, 1987). kamoto et al., 2003; Usami et al., 1999). Not all cases of The altered ABR phenotype of the mutants correlated non-syndromic EVA were accounted for by mono- and with a marked morphogenetic defect of the cochlea. Simi- diallelic mutations in SLC26A4 suggesting that mutations larly truncated cochleae were observed in mutant alleles in other loci may cause non-syndromic EVA or modify of genes that are known to be involved in patterning path- mutant SLC26A4 alleles (Pryor et al., 2005; Tsukamoto ways such as Pax2, Eya1, Ltap1,andDvl2 (Curtin et al., et al., 2003; Usami et al., 1999). The Slc12a2 gene, which 2003; Johnson et al., 1999; Montcouquiol et al., 2003; Tor- shows motif similarities to Slc26a4 and also underlies
Table 5 Summary of the mapping statistic Cross Trait Marker LOD % CI Add B6/CZECH-jc/jc N2 8 kHz D18Mit60 3.2 29 43 À22.71 16 kHz D18Mit60 3.5 32 40 À27.99 B6/CZECH-jc/jc F2 8 kHz D9Mit151 5.5 43 27 À14.49 16 kHz D9Mit151 4.8 39 30 À13.62 %, Percent variation explained by marker locus; CI, 95% percent confidence interval; regression coefficient. 56 A. Calderon et al. / Hearing Research 221 (2006) 44–58 deafness in the shaker-with-syndactylism mutant (syfp)isa 4.3. Hearing thresholds and genetic background good candidate for such a second locus, but jc might also be involved in EVA and Mondini dysplasia (Delpire Expression of the Jackson circler phenotype was highly et al., 1999; Dixon et al., 1999; Flagella et al., 1999). How- dependent upon the genetic background. A significant var- ever, despite the gross-morphological similarities with iability of hearing thresholds was already observed on the Mondini dysplasia, our paint-filling studies did not reveal C57BL/6J isogenic strain, which may indicate that jc is not a significantly enlarged endolymphatic duct in the mutants a complete loss-of-function allele. On both the CAST/Ei at E15.5. Instead, compared with the dilated endolym- and CZECHII/Ei background, hearing thresholds in jc phatic duct and vestibule at E15.5-16.5 seen in Slc26a4- homozygotes were widely distributed, and in some deficient mice (Everett et al., 2001), the membranous laby- instances the hearing loss was completely rescued. The rinth in jc mutants appeared fairly normal. It is possible large threshold shift towards the normal hearing range in however that a dilation or enlargement of the endolym- the F2 and N2 progeny and the differing distribution in phatic duct in jc mutants may occur later in development. the backcross versus the intercross strongly argue that genetic factors control this distribution. In the B6/ 4.2. Erratic circling behavior CZECH-jc backcross and B6/CZECH-jc intercross, signif- icant linkage was obtained with markers on chromosome Ambulatory and circling behavior of jc mice was ascer- 18 and 9, respectively. However these marker alleles were tained by quantifying movements in an open field. Consis- inversely correlated with hearing thresholds in both tent with the previous observation that circling becomes crosses, such that lower thresholds were associated with more pronounced with age (Southard, 1970) we found C57BL/6J-derived alleles and not with CZECH-alleles as that three to five week-old homozygotes with normal expected. In addition, phenotype-based backcrossing was ambulation times developed abnormal movement behav- unsuccessful in isolating a genomic segment that could be ior by six weeks of age. As expected for a behavioral trait, linked unambiguously with better hearing, although we we also found a wide scatter of data points and means were able to clearly identify and segregate the trait. While with large standard deviations. Despite this variability, many quantitative traits can be mapped with genome-wide there were clear differences between the test groups that significant LOD scores, isolating the congenic intervals included common laboratory and wild-derived strains and identifying the quantitative trait nucleotide(s) (QTNs) and jc homozygotes. Jones and colleagues recently has proven to be much more difficult (Flint et al., 2005; assessed gravity receptor function by measuring vestibu- Legare et al., 2000). In cases where hearing loss QTNs lar-evoked potentials (VsEPs) in a series of mutants and hearing modifiers were molecularly identified, the including the Jackson circler (Jones et al., 2005). No locus accounted for a significant portion of the overall var- response could be elicited in six-week old jc mice (n =5) iance and could also be interpreted as a Mendelian trait suggesting a pathology in sensory hair cells of saccule with reduced penetrance (Ikeda et al., 1999; Noben-Trauth and utricle (Jones et al., 2005). et al., 1997). In our study, it is possible that numerous To further distinguish the vestibular abnormality, a small-effect modifier genes controlled the variation in the panel of circling mutants was included in the study. All F2 and N2 crosses, which could not be detected due to of the panel’s mutants had a known pathology in the ves- the limited sample size. The lower thresholds in the tibular neuroepithelium (Deol, 1954; Di Palma et al., advanced backcross generations, are probably due to sto- 2001; Sjo¨strom and Anniko, 1990), with the exception of chastic effects, which we also observed in jc homozygotes Pcdh15av (Alagramam et al., 2005). Under our test condi- in the C57BL/6J background. tions, ambulation time and circling behavior was indistin- guishable among the different mutant strains. Clearly, Acknowledgements this was due, at least in part, to the presence of similar defects in the vestibule. Considered together with the The authors thank Melissa Irby and Jimmy Fiallos for results obtained by Jones and colleagues these data suggest technical assistance, ABR measurements, open-field test- that a primary defect in the vestibular neuroepithelium was ing, and maintenance of the mouse colonies. We thank the cause of the circling behavior. Interestingly, Lo¨scher Doris Wu for help with the paint fillings. We thank Doris and colleagues recently found that altered rate and dis- Wu and Matt Kelley for reading the manuscript. This work charge patterns in basal ganglia neurons correlated with was supported by the Intramural Research Program at circling and hyperactivity behavior in the ci2 rat mutant NIDCD/NIH (K.N.T.), DC000613 and DC003114 to (Fedrowitz et al., 2003; Kaiser et al., 2001). Besides the cen- GKM and DC04301 to K.R.J. tral and behavioral phenotypes, ci2 rats are deaf due to degeneration of the organ of Corti but show a normal ves- References tibular neuroepithelium, similar to the Ames waltzer mouse mutant (Alagramam et al., 2005). It may be that likewise Alagramam, K.N., Stahl, J.S., Jones, S.M., Pawlowski, K.S., Wright, affected central pathways contribute to the abnormal C.G., 2005. Characterization of vestibular dysfunction in the mouse movement behavior in jc mice. model for Usher syndrome 1F. J. Assoc. Res. Otolaryngol. 6, 106–118. A. Calderon et al. / Hearing Research 221 (2006) 44–58 57
Campbell, C., Cucci, R.A., Prasad, S., Green, G.E., Edeal, J.B., Galer, Hunter, K.P., Willott, J.F., 1987. Aging and the auditory brainstem C.E., Karniski, L.P., Sheffield, V.C., Smith, R.J., 2001. Pendred response in mice with severe or minimal presbycusis. Hear. Res. 30, syndrome, DFNB4, and PDS/SLC26A4 identification of eight novel 207–218. mutations and possible genotype-phenotype correlations. Hum. Ikeda, A., Zheng, Q.Y., Rosenstiel, P., Maddatu, T., Zuberi, A.R., Mutat. 17, 403–411. Roopenian, D.C., North, M.A., Naggert, J.K., Johnson, K.R., Churchill, G.A., Doerge, R.W., 1994. Empirical threshold values for Nishina, P.M., 1999. Genetic modification of hearing in tubby mice: quantitative trait mapping. Genetics 138, 963–971. evidence for the existence of a major gene (moth1), which protects Cordes, S.P., Barsh, G.S., 1994. The mouse segmentation gene kr encodes tubby mice from hearing loss. Hum. Mol. Genet. 8, 1761–1767. a novel basic domain-leucine zipper transcription factor. Cell 79, 1025– Iughetti, P., Alonso, L.G., Wilcox, W., Alonso, N., Passos-Bueno, M.R., 1034. 2000. Mapping of the autosomal recessive (AR) craniometaphyseal Curtin, J.A., Quint, E., Tsipouri, V., Arkell, R.M., Cattanach, B., Copp, dysplasia locus to chromosome region 6q21-22 and confirmation of A.J., Henderson, D.J., Spurr, N., Stanier, P., Fisher, E.M., Nolan, genetic heterogeneity for mild AR spondylocostal dysplasia. Am. J. P.M., Steel, K.P., Brown, S.D., Gray, I.C., Murdoch, J.N., 2003. Med. Genet. 95, 482–491. Mutation of Celsr1 disrupts planar polarity of inner ear hair cells and Jimenez, A.M., Stagner, B.B., Martin, G.K., Lonsbury-Martin, B.L., causes severe neural tube defects in the mouse. Curr. Biol. 13, 1129– 1999. Age-related loss of distortion product otoacoustic emissions in 1133. four mouse strains. Hear. Res. 138, 91–105. Delpire, E., Lu, J., England, R., Dull, C., Thorne, T., 1999. Deafness and Jimenez, A.M., Stagner, B.B., Martin, G.K., Lonsbury-Martin, B.L., imbalance associated with inactivation of the secretory Na-K-2Cl co- 2001. Susceptibility of DPOAEs to sound overexposure in inbred mice transporter. Nat. Genet. 22, 192–195. with AHL. J. Assoc. Res. Otolaryngol. 2, 233–245. Deol, M.S., 1954. The anomalies of the labyrinth of the mutants Johnson, K.R., Cook, S.A., Erway, L.C., Matthews, A.N., Sanford, L.P., varitint-waddler, shaker-2 and jerker in the mouse. J. Genet. 52, Paradies, N.E., Friedman, R.A., 1999. Inner ear and kidney anomalies 562–588. caused by IAP insertion in an intron of the Eya1 gene in a mouse Di Palma, F., Holme, R.H., Bryda, E.C., Belyantseva, I.A., Pellegrino, R., model of BOR syndrome. Hum. Mol. Genet. 8, 645–653. Kachar, B., Steel, K.P., Noben-Trauth, K., 2001. Mutations in Cdh23, Johnson, K.R., Zheng, Q.Y., Noben-Trauth, K., 2006. Strain background encoding a new type of cadherin, cause stereocilia disorganization in effects and genetic modifiers of hearing in mice. Brain Res. 1091, 79–88. waltzer, the mouse model for Usher syndrome type 1D. Nat. Genet. Jones, S.M., Johnson, K.R., Yu, H., Erway, L.C., Alagramam, K.N., 27, 103–107. Pollak, N., Jones, T.A., 2005. A quantitative survey of gravity receptor Dixon, M.J., Gazzard, J., Chaudhry, S.S., Sampson, N., Schulte, function in mutant mouse strains. J. Assoc. Res. Otolaryngol., 1–14. B.A., Steel, K.P., 1999. Mutation of the Na–K–Cl co-transporter Kaiser, A., Fedrowitz, M., Ebert, U., Zimmermann, E., Hedrich, H.J., gene Slc12a2 results in deafness in mice. Hum. Mol. Genet. 8, Wedekind, D., Loscher, W., 2001. Auditory and vestibular defects in 1579–1584. the circling (ci2) rat mutant. Eur. J. Neurosci. 14, 1129–1142. Eppig, J.T., Bult, C.J., Kadin, J.A., Richardson, J.E., Blake, J.A., Keithley, E.M., Canto, C., Zheng, Q.Y., Fischel-Ghodsian, N., Johnson, Anagnostopoulos, A., Baldarelli, R.M., Baya, M., Beal, J.S., Bello, K.R., 2004. Age-related hearing loss and the ahl locus in mice. Hear. S.M., Boddy, W.J., Bradt, D.W., Burkart, D.L., Butler, N.E., Res. 188, 21–28. Campbell, J., Cassell, M.A., Corbani, L.E., Cousins, S.L., Dahmen, Lander, E., Kruglyak, L., 1995. Genetic dissection of complex traits: D.J., Dene, H., Diehl, A.D., Drabkin, H.J., Frazer, K.S., Frost, P., guidelines for interpreting and reporting linkage results. Nat. Genet. Glass, L.H., Goldsmith, C.W., Grant, P.L., Lennon-Pierce, M., Lewis, 11, 241–247. J., Lu, I., Maltais, L.J., McAndrews-Hill, M., McClellan, L., Miers, Legare, M.E., Bartlett 2nd, F.S., Frankel, W.N., 2000. A major effect QTL D.B., Miller, L.A., Ni, L., Ormsby, J.E., Qi, D., Reddy, T.B., Reed, determined by multiple genes in epileptic EL mice. Genome Res. 10, D.J., Richards-Smith, B., Shaw, D.R., Sinclair, R., Smith, C.L., 42–48. Szauter, P., Walker, M.B., Walton, D.O., Washburn, L.L., Witham, Manly, K.F., Cudmore Jr., R.H., Meer, J.M., 2001. Map Manager QTX, I.T., Zhu, Y., 2005. The mouse genome database (MGD): from genes cross-platform software for genetic mapping. Mamm. Genome 12, to mice – a community resource for mouse biology. Nucleic Acids Res. 930–932. 33, D471–D475. Martin, P., Swanson, G.J., 1993. Descriptive and experimental analysis of Everett, L.A., Belyantseva, I.A., Noben-Trauth, K., Cantos, R., Chen, A., the epithelial remodellings that control semicircular canal formation in Thakkar, S.I., Hoogstraten-Miller, S.L., Kachar, B., Wu, D.K., Green, the developing mouse inner ear. Dev. Biol. 159, 549–558. E.D., 2001. Targeted disruption of mouse Pds provides insight about Mburu, P., Mustapha, M., Varela, A., Weil, D., El-Amraoui, A., Holme, the inner-ear defects encountered in Pendred syndrome. Hum. Mol. R.H., Rump, A., Hardisty, R.E., Blanchard, S., Coimbra, R.S., Genet. 10, 153–161. Perfettini, I., Parkinson, N., Mallon, A.M., Glenister, P., Rogers, Fedrowitz, M., Lindemann, S., Loscher, W., Gernert, M., 2003. Altered M.J., Paige, A.J., Moir, L., Clay, J., Rosenthal, A., Liu, X.Z., Blanco, spontaneous discharge rate and pattern of basal ganglia output G., Steel, K.P., Petit, C., Brown, S.D., 2003. Defects in whirlin, a PDZ neurons in the circling (ci2) rat mutant. Neuroscience 118, 867–878. domain molecule involved in stereocilia elongation, cause deafness in the Flagella, M., Clarke, L.L., Miller, M.L., Erway, L.C., Giannella, R.A., whirler mouse and families with DFNB31. Nat. Genet. 34, 421–428. Andringa, A., Gawenis, L.R., Kramer, J., Duffy, J.J., Doetschman, T., Melcher, J.R., Kiang, N.Y., 1996a. Generators of the brainstem auditory Lorenz, J.N., Yamoah, E.N., Cardell, E.L., Shull, G.E., 1999. Mice evoked potential in cat. III: Identified cell populations. Hear. Res. 93, lacking the basolateral Na-K-2Cl cotransporter have impaired epithe- 52–71. lial chloride secretion and are profoundly deaf. J. Biol. Chem. 274, Melcher, J.R., Guinan Jr., J.J., Knudson, I.M., Kiang, N.Y., 1996b. 26946–26955. Generators of the brainstem auditory evoked potential in cat. II. Flint, J., Valdar, W., Shifman, S., Mott, R., 2005. Strategies for mapping Correlating lesion sites with waveform changes. Hear. Res. 93, 28–51. and cloning quantitative trait genes in rodents. Nat. Rev. Genet. 6, Melcher, J.R., Knudson, I.M., Fullerton, B.C., Guinan Jr., J.J., Norris, 271–286. B.E., Kiang, N.Y., 1996c. Generators of the brainstem auditory Hasson, T., Heintzelman, M.B., Santos, S.J., Corey, D.P., Mooseker, evoked potential in cat. I. An experimental approach to their M.S., 1995. Expression in cochlea and retina of myosin VIIa, the gene identification. Hear. Res. 93, 1–27. product defective in Usher syndrome type 1B. Proc. Natl. Acad. Sci. MGD. 2005. Mouse Genome Informatics Project, The Jackson Labora- USA 92, 9815–9819. tory, Bar Harbor, ME. World Wide Web (URL:
Montcouquiol, M., Rachel, R.A., Lanford, P.J., Copeland, N.G., Jenkins, Tsukamoto, K., Suzuki, H., Harada, D., Namba, A., Abe, S., Usami, S., N.A., Kelley, M.W., 2003. Identification of Vangl2 and Scrb1 as 2003. Distribution and frequencies of PDS (SLC26A4) mutations in planar polarity genes in mammals. Nature 423, 173–177. Pendred syndrome and nonsyndromic hearing loss associated with Mundinus, C., 1791. Opuscula Caroli Mundini: Anatomica surdi natio enlarged vestibular aqueduct: a unique spectrum of mutations in sectio. De Bononiensi Scientarium et Artium atque Academia Com- Japanese. Eur. J. Hum. Genet. 11, 916–922. mentarii. 7. Usami, S., Abe, S., Weston, M.D., Shinkawa, H., Van Camp, G., Nadeau, J.H., 2001. Modifier genes in mice and humans. Nat. Rev. Genet. Kimberling, W.J., 1999. Non-syndromic hearing loss associated with 2, 165–174. enlarged vestibular aqueduct is caused by PDS mutations. Hum. Nadeau, J.H., 2003. Modifier genes and protective alleles in humans and Genet. 104, 188–192. mice. Curr. Opin. Genet. Dev. 13, 290–295. Vazquez, A.E., Luebke, A.E., Martin, G.K., Lonsbury-Martin, B.L., Noben-Trauth, K., Zheng, Q.Y., Johnson, K.R., Nishina, P.M., 1997. 2001. Temporary and permanent noise-induced changes in distortion mdfw: a deafness susceptibility locus that interacts with deaf waddler product otoacoustic emissions in CBA/CaJ mice. Hear. Res. 156, 31– (dfw). Genomics 44, 266–272. 43. Phippard, D., Lu, L., Lee, D., Saunders, J.C., Crenshaw 3rd, E.B., 1999. Wang, J., Mark, S., Zhang, X., Qian, D., Yoo, S.J., Radde-Gallwitz, K., Targeted mutagenesis of the POU-domain gene Brn4/Pou3f4 causes Zhang, Y., Lin, X., Collazo, A., Wynshaw-Boris, A., Chen, P., 2005. developmental defects in the inner ear. J. Neurosci. 19, 5980–5989. Regulation of polarized extension and planar cell polarity in the Pryor, S.P., Madeo, A.C., Reynolds, J.C., Sarlis, N.J., Arnos, K.S., cochlea by the vertebrate PCP pathway. Nat. Genet. 37, 980–985. Nance, W.E., Yang, Y., Zalewski, C.K., Brewer, C.C., Butman, J.A., Witmer, P.D., Doheny, K.F., Adams, M.K., Boehm, C.D., Dizon, J.S., Griffith, A.J., 2005. SLC26A4/PDS genotype-phenotype correlation in Goldstein, J.L., Templeton, T.M., Wheaton, A.M., Dong, P.N., Pugh, hearing loss with enlargement of the vestibular aqueduct (EVA): E.W., Nussbaum, R.L., Hunter, K., Kelmenson, J.A., Rowe, L.B., evidence that Pendred syndrome and non-syndromic EVA are distinct Brownstein, M.J., 2003. The development of a highly informative clinical and genetic entities. J. Med. Genet. 42, 159–165. mouse simple sequence length polymorphism (SSLP) marker set and Quint, E., Steel, K., 2003. Use of mouse genetics for studying inner ear construction of a mouse family tree using parsimony analysis. Genome development. Curr. Top. Dev. Biol. 57, 45–83. Res. 13, 485–491. Rzadzinska, A.K., Schneider, M.E., Davies, C., Riordan, G.P., Kachar, Xu, P.X., Adams, J., Peters, H., Brown, M.C., Heaney, S., Maas, R., B., 2004. An actin molecular treadmill and myosins maintain stereo- 1999. Eya1-deficient mice lack ears and kidneys and show abnormal cilia functional architecture and self-renewal. J. Cell Biol. 164, 887– apoptosis of organ primordia. Nat. Genet. 23, 113–117. 897. Zheng, Q.Y., Johnson, K.R., Erway, L.C., 1999. Assessment of hearing in Siemens, J., Kazmierczak, P., Reynolds, A., Sticker, M., Littlewood- 80 inbred strains of mice by ABR threshold analyses. Hear. Res. 130, Evans, A., Muller, U., 2002. The Usher syndrome proteins cadherin 23 94–107. and harmonin form a complex by means of PDZ-domain interactions. Zheng, L., Sekerkova, G., Vranich, K., Tilney, L.G., Mugnaini, E., Proc. Natl. Acad. Sci. USA 99, 14946–14951. Bartles, J.R., 2000. The deaf jerker mouse has a mutation in the gene Sjo¨strom, B., Anniko, M., 1990. Morphologically specific vestibular hair encoding the espin actin-bundling proteins of hair cell stereocilia and cell degeneration in the jerker mouse mutant. Eur. Arch. Otorhino- lacks espins. Cell 102, 377–385. laryngol. 247, 51–55. Zhou, R., Abbas, P.J., Assouline, J.G., 1995a. Electrically evoked Southard, J.L., 1970. Jackson circler, jc. Mouse News Lett. 42, 30. auditory brainstem response in peripherally myelin-deficient mice. Steel, K.P., Kros, C.J., 2001. A genetic approach to understanding Hear. Res. 88, 98–106. auditory function. Nat. Genet. 27, 143–149. Zhou, R., Assouline, J.G., Abbas, P.J., Messing, A., Gantz, B.J., 1995b. Torres, M., Gomez-Pardo, E., Gruss, P., 1996. Pax2 contributes to inner ear Anatomical and physiological measures of auditory system in mice patterning and optic nerve trajectory. Development 122, 3381–3391. with peripheral myelin deficiency. Hear. Res. 88, 87–97.